open micro pmu: a€¦ · open micro‐pmu: a real world ... • immense time series data base...

16
Open MicroPMU: A real world reference distribution microphasor measurement unit data set for research and application development Emma M. Stewart – Lawrence Livermore National Laboratory July 18 th 2017 IEEE PES GM: Big Data Access and Big Data Research Integration in Power Systems 1

Upload: others

Post on 18-Oct-2020

0 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Open Micro PMU: A€¦ · Open Micro‐PMU: A real world ... • Immense time series data base bandwidth in the cloud – 60M inserts per second (ips), 120M queries per second –

Open Micro‐PMU: A real world reference distribution micro‐phasor 

measurement unit data set for research and application development

Emma M. Stewart – Lawrence Livermore National Laboratory

July 18th 2017 IEEE PES GM: Big Data Access and Big Data Research 

Integration in Power Systems

1

Page 2: Open Micro PMU: A€¦ · Open Micro‐PMU: A real world ... • Immense time series data base bandwidth in the cloud – 60M inserts per second (ips), 120M queries per second –

Overview

• microPMUs – how we got here, where we are going

• Powerdata.lbl.gov• BTrDB• Use Cases for microPMU data• Contact information 

2

Page 3: Open Micro PMU: A€¦ · Open Micro‐PMU: A real world ... • Immense time series data base bandwidth in the cloud – 60M inserts per second (ips), 120M queries per second –

ARPA‐ funded project which developed a network of high‐precision phasor measurement units (µPMUs) to measure voltage phase angle to within 0.01o

Built Berkeley Tree Database (BTrDB) for collecting raw data at 120 Hz, processing ‘distillate’ streams online, visualizing and archiving dataGetting µPMU data into distribution systems:

• Transitioning to commercial systems• Sold as a quick start kit from PSL with a server and 

BTrDB• 40+ are installed in the field (through this 

research, 100’s more now)• Utilize combination of 4G LTE, • Possible Locations: Piggy backed onto revenue 

metering, where exiting PT and CT locations or at simple plug in customer locations for single phase information

Background to Micro‐synchrophasors (µPMUs) for Distribution Systems

Page 4: Open Micro PMU: A€¦ · Open Micro‐PMU: A real world ... • Immense time series data base bandwidth in the cloud – 60M inserts per second (ips), 120M queries per second –

Reference Dataset

LBNL is making a 3-month reference dataset from three micro-PMUs on a 12-kV system available for researchersAvailable through 3 mediums…BTrDB (RHS)Or a bulk download

see powerdata.lbl.gov

Powerdata.lbl.gov a reference microPMUdataset

Download or visualize in

BTrDB

Bulk download

Real time data from an LBNL

building

Page 5: Open Micro PMU: A€¦ · Open Micro‐PMU: A real world ... • Immense time series data base bandwidth in the cloud – 60M inserts per second (ips), 120M queries per second –

BTrDB Interface for PowerData Set5

Login

Short section download

Select times

Multiple axes Scalable graph

Page 6: Open Micro PMU: A€¦ · Open Micro‐PMU: A real world ... • Immense time series data base bandwidth in the cloud – 60M inserts per second (ips), 120M queries per second –

LBNL Installation (1) • 3 uPMUs in 1 electrical path• Oct 1 2015 to Dec 31 2015

– Normal load at the lab and shutdown period from Dec 24 – 31

• All locations for the data these measurements have been scaled to the primary voltage with the applied PT and CT ratios. 

• LBNL 1 (Grizzly bus1 2): – 7.2kV‐120V x 3 0.3 Class PTs – 1.2 Class CT 1200:5A, PSL precision CT 5A: – 0.333mV 

• There are two copper busbars of negligible impedance between LBNL 1 and 2 and the following underground cable. 

• 6‐750 KCMIL, underground cable, 2459 ft. 

6

Page 7: Open Micro PMU: A€¦ · Open Micro‐PMU: A real world ... • Immense time series data base bandwidth in the cloud – 60M inserts per second (ips), 120M queries per second –

LBNL Installation (2)• LBNL 2 (a6 bus1): 

– 7.2kV‐120V x 3 0.3 Class PTs – 1.2 Class CT 1200:5A, PSL precision CT – 5A:0.333mV 

• LBNL 3 (bank 514):– 480V/208VL‐L– 1.2 Class CT 9600:1A, PSL precision CT 5A:0.333mV

• LBNL 3 is connected on the low side of a distribution transformer with the following specification 

– Delta Y‐G1500/2000 KVA, AA/FA, 12.47kV/480Y‐277V, 

– Z=5.75%

7

Page 8: Open Micro PMU: A€¦ · Open Micro‐PMU: A real world ... • Immense time series data base bandwidth in the cloud – 60M inserts per second (ips), 120M queries per second –

BTrDB – Massive Energy System Analytics

● Archiver / Database● Stores (T, V) pairs● Nanosecond precision● Fault tolerant● Highly scalable● Unique abstraction

● query range (ver)● insert values => ver● delete range => ver● query statistical (ver)● compute diff(v1, v2)

SCE Riverside

Alabama Georgia

Berkeley LBNL Alameda PG&E LBNL/CEC LBNL/NSA

Tennessee

DARPA Pepco

Sandia

Navy Yard

http://btrdb.io/

Michael Anderson ARPA- E Slide

Page 9: Open Micro PMU: A€¦ · Open Micro‐PMU: A real world ... • Immense time series data base bandwidth in the cloud – 60M inserts per second (ips), 120M queries per second –

Four Transformative Advances• Immense time series data base bandwidth in the cloud

– 60M inserts per second (ips), 120M queries per second– All 1,700 PMUs in North America generate 2M ips– 300M smart meters generate 0.33M ips– Stream ALL the data to the cloud, innovate on analytics

• Distillation infrastructure with extremely fast change set identification– Operate real‐time on many streams, with holes, out‐of‐order, etc.

• On‐the‐Fly statistical summaries– Over a multi‐resolution store

• Multi‐resolution search and process– Find ‘needle’ events in immense haystacks instantly– Drill down exponentially to analyze– WITH algorithm innovation at the core

• Now in commercial use with 

Page 10: Open Micro PMU: A€¦ · Open Micro‐PMU: A real world ... • Immense time series data base bandwidth in the cloud – 60M inserts per second (ips), 120M queries per second –

Time scales of measurement and operations are diverse

Key Questions for both research and industryWhat measurements do we need?

Where do we need the measurement and analytics?Who needs the information?

When and How do they need it

Useful Actionable Information

Useful Analyzable Data

Page 11: Open Micro PMU: A€¦ · Open Micro‐PMU: A real world ... • Immense time series data base bandwidth in the cloud – 60M inserts per second (ips), 120M queries per second –

Proven Use Cases on Real Data

• Large PV Site Behavior Tracking – State of the PV report – impacts analysis• Disaggregation of behind the meter generation (with one sensor)• Load, Impedance and PV Model validation for distribution planning• High Impedance Fault Detection• Advance prediction of transformer/equipment failure• Future: Control and DER Management system integration for ops 

Page 12: Open Micro PMU: A€¦ · Open Micro‐PMU: A real world ... • Immense time series data base bandwidth in the cloud – 60M inserts per second (ips), 120M queries per second –

Voltage sag observed at Northern and Southern CA locations – same event 500 miles apart

Sout

hern

CA

Sout

hern

CA

Use Cases: Locating both system and local events

Page 13: Open Micro PMU: A€¦ · Open Micro‐PMU: A real world ... • Immense time series data base bandwidth in the cloud – 60M inserts per second (ips), 120M queries per second –

Incipient Failure Detection for Transformers (and other equipment)

• In the US – transformers are in general a big point of failure in the aging distribution system –when the fail, they cause an outage and $$ to replace 

• Application picked up the signature below multiple times– Tap change followed by voltage sag – multiple times– We can only see this relational information with synchronized datasets from the uPMU– Tap changer oil leak – signature is evident before normal warning of failure

Tap changer close to failure

Page 14: Open Micro PMU: A€¦ · Open Micro‐PMU: A real world ... • Immense time series data base bandwidth in the cloud – 60M inserts per second (ips), 120M queries per second –

Summary• Data Citation: Stewart E.M., et al "Open μPMU: A real world reference 

distribution micro‐phasor measurement unit data set for research and application development," LBNL Technical Report 1006408, October 2016.

• Other publications:– Stewart E.M., M. Stadler, C. Roberts, J. Reilly, D. Arnold, J. Joo, “Data Driven Approach for 

Monitoring Protection and Control of Distribution System Assets using MicroPMUtechnology” CIRED 2017, Glasgow Scotland

– Alexandra von Meier⇤, Emma Stewart†, Alex McEachern‡, Michael Andersen§, and Laura Mehrmanesh, 2017, Precision Micro‐Synchrophasors for Distribution Systems: A Summary of Applications, IEEE Transactions on Smart Grid

– Mahdi Jamei, Emma Stewart, Ciaran Roberts, Alex McEachern, Anna Scaglione, Sean Peisert, Online Thevenin Parameter Tracking Using Synchrophasor Data, IEEE PES GM 2017

14

Page 15: Open Micro PMU: A€¦ · Open Micro‐PMU: A real world ... • Immense time series data base bandwidth in the cloud – 60M inserts per second (ips), 120M queries per second –

Contact Information

• Emma M Stewart (general questions): LLNL  stewart [email protected] 

• Michael Andersen (BTrDB): [email protected]

• Sean Peisert (powerdata specific questions): LBNL, [email protected]

15

Page 16: Open Micro PMU: A€¦ · Open Micro‐PMU: A real world ... • Immense time series data base bandwidth in the cloud – 60M inserts per second (ips), 120M queries per second –

Micro-synchrophasors (µPMUs)for disibution systems

Three-year, $4.4 M ARPA-E OPEN 2012 project (2013-2016)Research partners CIEE, UC Berkeley, Lawrence Berkeley Lab, Power Standards Lab

Field installations at Riverside Public Utilities, Southern California Edison, Pacific Gas & Electric, Alabama Power, Georgia Power, Tennessee Valley Authority